nanog mailing list archives
Re: buffer bloat and packet pacing
From: Nick Hilliard <nick () foobar org>
Date: Thu, 3 Sep 2015 13:04:34 +0100
On 03/09/2015 11:56, Saku Ytti wrote:
40GE server will flood the window as fast as it can, instead of limiting itself to 10Gbps, optimally it'll send at linerate.
optimally, but tcp slow start will generally stop this from happening on well behaved sending-side stacks so you send up ramping up quickly to path rate rather than egress line rate from the sender side. Also, regardless of an individual flow's buffering requirements, the intermediate path will be catering with large numbers of flows, so while it's interesting to talk about 375mb of intermediate path buffers, this is shared buffer space and any attempt on the part of an individual sender to (ab)use the entire path buffer will end up causing RED/WRED for everyone else. Otherwise, this would be a fascinating talk if people had real world data. Nick
Current thread:
- buffer bloat and packet pacing Saku Ytti (Sep 03)
- Re: buffer bloat and packet pacing Saku Ytti (Sep 03)
- Re: buffer bloat and packet pacing Nick Hilliard (Sep 03)
- Re: buffer bloat and packet pacing Saku Ytti (Sep 03)
- Re: buffer bloat and packet pacing Brett Frankenberger (Sep 03)
- Re: buffer bloat and packet pacing Saku Ytti (Sep 03)
- Re: buffer bloat and packet pacing Brett Frankenberger (Sep 03)
- Software Defined Networking Rod Beck (Sep 04)
- Re: Software Defined Networking Jethro R Binks (Sep 04)
- Re: Software Defined Networking Niraj Kacha (Sep 04)
- Re: Software Defined Networking Narseo Vallina Rodriguez (Sep 05)
- Re: Software Defined Networking John Kristoff (Sep 04)
- Message not available
- Re: Software Defined Networking Ignacio de castro (Sep 05)